Elsevier

Journal of Chromatography A

Volume 1229, 16 March 2012, Pages 208-215
Journal of Chromatography A

Development of a hydrophilic interaction liquid chromatography–mass spectrometry method for detection and quantification of urea thermal decomposition by-products in emission from diesel engine employing selective catalytic reduction technology

https://doi.org/10.1016/j.chroma.2012.01.046Get rights and content

Abstract

The use of urea based selective catalytic reduction (SCR) technology for the reduction of NOx from the exhaust of diesel-powered vehicles has the potential to emit at least six thermal decomposition by-products, ammonia, and unreacted urea from the tailpipe. These compounds may include: biuret, dicyandiamine, cyanuric acid, ammelide, ammeline and melamine. In the present study, a simple, sensitive and reliable hydrophilic interaction liquid chromatography (HILIC)-electrospray ionization (ESI)/mass spectrometry (MS) method without complex sample pre-treatment was developed for identification and determination of urea decomposition by-products in diesel exhaust. Gradient separation was performed on a SeQuant® ZIC-HILIC column with a highly polar zwitterionic stationary phase, and using a mobile phase consisting of acetonitrile (eluent A) and 15 mM ammonium formate (pH 6; eluent B). Detection and quantification were performed using a quadrupole ESI/MS operated simultaneously in negative and positive mode. With 10 μL injection volume, LODs for all target analytes were in the range of 0.2–3 μg/L. The method showed a good inter-day precision of retention time (RSD < 0.5%) and peak area (RSD < 3%). Satisfactory extraction recoveries from spiked blanks ranged between 96 and 98%. Analyses of samples collected during transient chassis dynamometer tests of a bus engine equipped with a diesel particulate filter (DPF) and urea based SCR technology showed the presence of five target analytes with cyanuric acid and ammelide the most abundant compounds in the exhaust.

Highlights

► Exhaust from a bus engine equipped with diesel particulate filter and urea-selective reduction technology. ► Identification and quantification of urea thermal decomposition compounds in diesel emission were studied. ► A new HILIC-ESI/MS method has been developed. ► Five target analytes, with cyanuric acid and ammelide the most abundant compounds, were determined.

Introduction

In response to health concerns, regulators have promulgated more stringent emissions standards for particulate matter (PM) and NOx from diesel engines. PM emissions are controlled in part by diesel particulate filters (DPFs) that incorporate either a passive or active strategy for regenerating the filter substrates. Passive DPFs use a combination of exhaust gas temperature (heat) and a catalyst to initiate the regeneration process while active DPFs use an outside source to provide the heat for regeneration. Urea-selective catalytic reduction (SCR) and NOx adsorber catalyst are emerging as the current leading contenders for NOx control in diesel engines. Urea-SCR technology can substitute the use of NH3, which poses problems related to its toxicity and has the advantage of being less sensitive to sulfur as compared the NOx absorber-type catalysts [1], [2], [3], [4], [5]. Nevertheless, it is important to understand the effects that engine and after-treatment technologies may have on the reduction or formation of the full spectrum of chemical species that are not currently identified.

Urea thermal decomposition reaction is a two-step process that includes the formation of ammonia, and isocyanic acid as an intermediate product [6], [7], [8]. Ammonia is then involved in several reactions, which ultimately lead to the denitrification of flue gas [1], [4], [8]. The isocyanic acid is very reactive and can initiate the formation of larger molecular weight compounds such as cyanuric acid, biuret, melamine, ammeline and ammelide [6], [7], [9], [10], [11]. Storey et al. [3] reported also the formation of dicyandiamide. However, the effect of urea-SCR technology in the chemical composition of the gaseous and PM diesel emissions is still ambiguous. In addition, the toxicity data base for these compounds, in general, is incomplete [12]. Thus, the quantification of these compounds is essential for identification of potential adverse environmental and health effects and to ensure that the emission inventories used in atmospheric modeling activities and in health risk assessments are accurate.

Various chromatographic approaches have already been developed and applied for the analysis of this class of chemicals in variety matrices. Previously, reversed-phase liquid chromatography (RPLC) with UV or diode array detection has been reported for the analysis of cyanuric acid in swimming pool waters and air [13], or melamine and its derivatives in natural waters and industrial products, and in air [14]. Recently, as an urgent response to toxicity incidents in 2007 and 2008 involving melamine [15], [16], [17] extensive efforts have been made for determining melamine and its analogs (ammeline, ammelide and cyanuric acid) in food/feed and animal tissues. In these incidents, the melamine was added deliberately to pet food and infant formula to boost their apparent protein content. Melamine alone is of low toxicity; however, combination with cyanuric acid leads to crystal formation and subsequent kidney toxicity [16]. Many novel analytical methods, based on gas chromatography–mass spectrometry (GC–MS) [18], [19], capillary electrophoresis with UV [20], [21], RPLC [22], [23] and hydrophilic interaction liquid chromatography (HILIC) with UV and MS detection [24], [25] have been developed. However, little is known about the determination of these highly polar nitrogenous compounds in emissions from diesel engines employing urea-SCR after-treatment technology. To our best knowledge, an anion-exchange chromatography with UV detection developed by Koebel and Elsner [9] is the only method reported to study the formation of thermal degradation by-products of the urea-SCR process.

The purpose of this work was to establish a reliable and sensitive method using LC-ESI/MS for identification and quantification of urea and its thermal decomposition by-products in emissions from diesel engine employing SCR technology. Recent advantages in LC–MS capabilities made this technique an attractive alternative to GC–MS because it offered the advantage of omitting a derivatization step. This was achieved by exploiting the advantages of RPLC and HILIC separations combined with ESI mass spectrometry giving molecular weight information. HILIC is a useful technique for the retention of polar analytes that offers a difference in selectivity compared to traditional RPLC [26], [27]. Different LC columns, which included a reversed-phase Zorbax SB-Aq and HILIC columns (viz., Atlantis™ HILIC Silica, Cogent Diamond Hydride™ and SeQuant® ZIC-HILIC), were tested. The best performance was obtained using the SeQuant® ZIC-HILIC column with a highly polar zwitterionic stationary phase. The developed HILIC-ESI/MS method was validated and successfully applied to the analysis of real-world samples following a simple sample pre-treatment procedure using acetonitrile/formic acid extraction.

Section snippets

Reagents and standards

Double deionized water (DDW; 18 MΩ; Barnstead, Dubuque, IA, USA) was used for the preparation of all solutions. HPLC grade acetonitrile (ACN), methanol and ammonium formate were purchased from Fisher Scientific (Ottawa, ON, Canada). Formic acid (98+%) was obtained from Acros Organics (Geel, Belgium). Cyanuric acid (CYA), melamine (MEL), dicyandiamide (DICY), biuret (BIU) and urea were purchased from Sigma–Aldrich (Toronto, ON, Canada). Ammeline (AML) was obtained from MP Biomedicals (Aurora, OH,

Optimization of the chromatographic separation

For LC separation of the target analytes listed in Table 1, the performance of different analytical columns with a reversed-phase and HILIC mode stationary phases were evaluated. Initially, a RPLC with a Zorbax SB-Aq column was tested (Fig. 1a). The Zorbax SB-Aq column was chosen as it was specifically designed to retain highly polar compounds allowing the use of highly aqueous mobile phases [28]. However, except for MEL, the target compounds were not well retained on the RPLC column under the

Conclusions

A new HILIC-ESI/MS method has been developed and validated for the identification and quantification of urea thermal decomposition compounds, namely cyanuric acid, biuret, dicyandiamide, ammeline, ammelide and melamine, in emissions from diesel engines employing urea-SCR after-treatment technology. Different RPLC and HILIC columns were tested. The best baseline separation of all analytes of interest was obtained using a highly polar zwitterionic stationary phase (SeQuant® ZIC-HILIC column) with

Acknowledgments

The authors appreciate financial support for this work from the Canadian Federal Program on Energy Research and Development (PERD) under the PERD C11.009 Project “Control and Reduction of Emissions of Particulate Matter”. The authors wish to gratefully thank Debbie Rosenblatt and Greg Rideout (Emissions Research and Measurement Section) for their helpful discussion during the study and providing samples.

References (38)

  • M. Koebel et al.

    Catal. Today

    (2000)
  • P.M. Schaber et al.

    Thermochim. Acta

    (2004)
  • O. Krocher et al.

    Anal. Chim. Acta

    (2005)
  • M. Koebel et al.

    J. Chromatogr. A

    (1995)
  • H.L. Fang et al.

    Appl. Catal. B

    (2003)
  • L. Stradella et al.

    Thermochim. Acta

    (1993)
  • I.-L. Tsai et al.

    J. Chromatogr. A

    (2009)
  • A.L.N. van Nuijs et al.

    J. Chromatogr. A

    (2011)
  • Y. Guo et al.

    J. Chromatogr. A

    (2011)
  • A. Alpert

    J. Chromatogr.

    (1990)
  • R.-I. Chirita et al.

    J. Chromatogr. A

    (2011)
  • W. Naidong

    J. Chromatogr. B

    (2003)
  • T. Ikegami et al.

    J. Chromatogr. A

    (2008)
  • W.H. Sun et al.
  • J.M.E. Storey et al.
  • C.S. Sluder et al.
  • J. Brouwer et al.
  • P.M. Schaber et al.

    Am. Lab.

    (1999)
  • J.C. Ball
  • Cited by (0)

    View full text